Biomedical Engineering Reference
In-Depth Information
process. The consequence of this approach is that scientists are required to
repeatedly switch from one application to another while pursuing a line of
thinking that requires access both to the literature and to data. Often the
situation is worse, and they are forced to print documents out, to scribble
notes in the margin and thereby remove the potential for interconnected
data completely. Not only does the act of consciously switching require a
certain amount of mental inertia (leading inevitably to, 'Oh, I'll look that
up next time when I'm using Application X'), it interrupts the workfl ow
and thought processes associated with reading the literature. In turn, this
makes documenting associations between concepts found in articles and
data recorded in other systems diffi cult and fragmented.
Consider a typical industry use-case: a scientist is reading a paper
describing a number of proteins and/or small molecules relevant to a disease
or phenotype. Some of these might be new to the reader, who may wish to
know a little more about them in order to better understand the fi ndings
in the paper. Traditionally, manual copy/pasting of protein and drug
names into numerous search engines is the only solution. For compounds,
this can also require tedious manual sketching of the molecule into a
structure editor. The diffi culty in performing what should be very
simple tasks often results in either lost time or simply abandoning the
enquiry.
Now, consider these same scenarios using Utopia Documents. Simply
selecting the protein name using the cursor brings up a rich information
panel about that entity (Figure 15.2). Alternatively, text-mining services
can be invoked directly from within the Utopia environment to
automatically identify interesting entities and relationships, and to link
these to in-house data. The result is a highly visual and organized index
of the genes, compounds, and diseases mentioned in the text. For
compounds, Utopia goes even further, as Figure 15.3 shows. Here, the
software has identifi ed a number of small molecules mentioned only by
name in the text of the paper. Using (in this example) the ChemSpider
[14] API, the actual structures have been retrieved, along with other
information, providing a completely new chemical index browser to the
paper. Functionality developed as part of the EU Open PHACTS project
[15] also allows the retrieval of activity data from public and corporate
structure-activity relationship databases.
Finally, because of Utopia's unique software architecture, many other
types of plug-in can be developed, including the ability to interact with
tables and to embed applications, such as 3D protein structure viewers,
directly in the document. Utopia provides a two-way API, allowing web
services to retrieve structured content from the PDF ('give me all tables
Search WWH ::
Custom Search